Background: Since October 2008, the Centers for Medicare and Medicaid Services (CMS) no longer pay hospitals for specific hospital-acquired conditions (HACs), regardless of patient risk factors for these conditions. However, changes in hospital payment considering risk-adjusted HAC rates are planned for 2015; yet, little is known about how HACs vary with patient characteristics available in claims data (such as age) that could be used readily for risk-adjustment. Physicians often consider elderly inpatients to be at higher risk for HACs.
Objective: To evaluate hospital billing abstracts for hospital-acquired condition cases, to describe how these complication rates may vary with advancing age.
Methods: We evaluated the administrative dataset for all adult discharges from acute care hospitals in the State of Michigan using the 2007 Healthcare Cost and Utilization Project State Inpatient Dataset. Potential HAC cases were identified using ICD-9-CM diagnosis and procedure codes specified in the CMS Hospital Acquired Conditions Initiative for FY2008, assuming all secondary diagnoses could be HACs because the variable to distinguish HACs from present-on-admission co-morbidities was not present prior to 2008. Age-specific rates were evaluated for 5 infections including urinary tract infections (UTIs), surgical site infections (SSIs) after hip or knee replacements or coronary artery bypass graft (CABG) surgery, and C. difficile and 2 non-infectious conditions of decubitus ulcers and blood clots (deep venous thrombosis/DVT, and pulmonary emboli/PE) after knee or hip replacement.
Results: Rates of HACs are compared by age category in the Table. The most common conditions that also increased with each age category included UTIs (increasing from 6.9% to 17.3%, though very few were identified as catheter-associated/CAUTIs) and decubitus ulcers (increasing from 1.6% to 3.7%). Neither orthopedic SSIs nor mediastinitis increased significantly with age, and very few were listed during the same admission as the hip or knee replacement or CABG procedure. Though not yet chosen by CMS for non-payment, C. difficile also increased with age, and patients with a secondary diagnosis of C.difficile often had another complication such as 31% with a UTI or CAUTI, and 14% with decubitus ulcer. Blood clots (DVT/PE) as secondary diagnoses occurred between 1.4-1.9% of all admissions at increased rates in patients >=65, with similar rates during admissions for hip or knee replacement.
Conclusions: Patients of advanced age are discharged with significantly higher rates of UTIs, decubitus ulcers, and C. difficile as secondary diagnoses, which may be hospital-acquired conditions. Age may be important in risk-adjustment for some HACs, which may be less preventable in patients of advanced age.